winz

I've been reading your AARs for some time, and they are fantastic! keep them coming :)

The campaign system works like this - Each campaign is divided into stages, and each stage has one, or more, missions assigned to it. The simulator then randomly picks one mission from the missions assigned to the stage you are currently in. After the you finnish your mission a new stage is picked based on your score. 0-49 - you move to the previous stage. 50 - you remain at your current stage, but another mission is picked. 51-100 - you move to the next stage. Narrated campaigns, like Deployment in BS/BS2, use only one mission per stage and never progress backwards (because the mission score is newer below 50). So you end up with the classic campaign scheme. Other campaigns, like the Georgian oil war campaign, have multiple mission per stage assigned and can progress both forward and backwards. You end up with a pseudo-dynamic campaign, that offer bigger replaybility.

Once again, wrong. Tessellation can be very beneficial for every engine that is trying to display huge areas. Why? First it's important to understand what tesselation really is. Tessellation is a geometry shader. Shader is a program that is run by the graphic card and can alter the scene in numerous ways. In the early days scenes were prety static, you loaded your models, textures into a scene and the GPU rendered it. If you wanted you change anything, you needed to redefine the scene (i.e. apply different texture to model), and this is very costly performance vise. Shader is a small program that allows you to change the scene (to a degree), without redefining the scene and can be used on tons of effects that would be impossible without them. Shaders are fast, because they are executed on the GPU. The most common shaders are vertex shader - one that can affect individual vertexes, and pixel shader - shader that can affect individual pixels. Geometry shader is a shader that affects, suprise surprise, geometry of an object. It allows you to write a small program, that will be executed by the GPU, that will change the polygon for the objects - it can add aditional vertexes. How can a sim engine profit from these capabilities? Simple, one thing sim engines strugle with is the variety of model fidelity they aim to reproduce. You want the ground to have nice details up-close, yet you don't need this detail when flying up-high. You want ground objects to be super-detailed up-close, yet you want to see them up far away. This is were LODs are used, but LODs aren't cost-free. Every LOD is a separate model, so, if you want 5 LODs for an object, you need 5 separate models, and you need to have these 5 separates models loaded in memory (even for ground). Also LOD transition (model pop-up) can be pretty disturbing if the differences between LODs are big (sure, you can smooth this out by making more LODs, but more LODs = more work, and more memory requirements). This is were tesselation can step in and help (it cannot replace LODs). You can write a shader that will increase the amount of detail on a object. You can smoothly increase the amount of detail as things gets closer, you can control the amount of detail added based on parameters (i.e. fps), you have huge control about what the shader is doing, because it's essentialy a program that alters the model. And you need only one model for every detail level. So tesselation can increase the ground fidelity a lot, it can increase the level of detail on those box houses, and it can do so smoothly.

Microsoft Flight cancelled. http://kotaku.com/5929141/microsoft-shuts-down-vancouver+based-studio I'm not suprised, that 'thing' was nor a flight sim, nor a game...and beeing a closed platform only sealed the deal.

/3GB has nothing to do with physical Ram. /3GB increases user-mode virtual memory to 3GB, while reducing kernel-mode memory to 1GB. By using /3GB you are reducing memory resources for kernel, drivers by half. Also many kernel internal structures are reduced by far more than 50%. This can cause problems.

Cannot wait for FC3... I sooo miss the stock Toad.

Do you actually own Black Shark 1? If not, then you downloaded the wrong file. You should download this, if you only own Black Shark 2 http://www.digitalcombatsimulator.com/index.php?scr=product&ProductId=12&end_pos=136&lang=en

What pressure? To release their source code? :D Well, that would certainly be an interesting business model :D

Because you would most likely need to modify the way camera/flir is rendered and for that you need access to the source code.

That's nonsense, human eye/brain aren't computers working in cycles, so speaking strictly in fps terms in relation to what is precieved as fluent motion, and what not, is nonsense. The image itself, sharpness, contrast..etc play a huge role. I.e. the eye will detect a totally white frame that is much shorter then 1/24 of a second. Yes, a movie appears fluent in 24fps, but that's because it's blurred, so the brain has bigger problem distinguish between individual frames, then in sharp pictures/movement generated by computer. And even in movies people have no problem to distinguish a 24fps movie from a 48fps one. I personally see the difference between 24fps and 60fps.

This might be a better fix, as the menuWidth change will not prevent the text to overflow to second monitor. The menu location is defined on line 103 :) menuPanel:setBounds(width - menuWidth, 0, menuWidth, height) Changing the first paramater to i.e 1600 menuPanel:setBounds(1600, 0, menuWidth, height) will set the menu on fix location 1600,0. This way, you will not get text overflow to second monitor. Note that if you changed the menuWidth on line 55, to move the menu, you need to change it back to it's default value, or your com menu will go FUBAR :) btw, thanks to Kuky for pointing me to the right file :)

You want to model engine operation, yet you don't want to use a modelling SW. The other option is even harder/much more expensive - CAT SW. Solidworks mentioned, or Catia, or similiar. Sorry, but I'm afraid 2d painting tools wont cut the deal in what you are trying to achieve.

It will not aks for activation key if you had BS2 previously installed and activated on your system.

Ofc, it isn't that hard for an educated person, but can be a pain for unexperienced PC user and it's harder to write down the instructions.

And then a patch will roll. Or, as we have now, integration with another product. And you lost all your bindings. Or you change the USB port you controller is plug into, you lost all settings. Or you have to reinstall your computer, or are migrating to a newer one, once a again, you lost all your settings. I personaly would recommend using the profiler, and not the in-game bindings. BTW, the profile you attached will not work for anyone else, but you. Why? Because the code in the { } is a OS generated UID for the USB device the profile is bound to. If you plug your controller to another usb the OS will generate another UID for it and DCS will thus recognize it as anothe controller and will not use the profile. If you plug the controller in another PC, the OS will, once again, assing a totally different UID..

Just a little hint. A good way to judge your approach speed is the AOA gauge. Small AOA - you're going too fast.

One simple thing you can do to make this a little less pita is to create two export files i.e. ExportA-10.lua, with helios enabled export for A-10, and ExportKa50.lua with helios enabled export for Ka-50. Then create two simple cmd files named ExportA-10.cmd copy /y ExportA-10.lua Export.lua and ExportKa50.cmd copy /y ExportKa50.lua Export.lua then place a shortcut for them on your desktop, and when you are about to fly with A-10 use the A-10 cmd file, and when Ka-50 use Ka-50 cmd.

I've no problem flying/landing/operating the Su-25T, are you sure the sim is the thing that's FUBAR?

AFAIK the Ka-50 uses doppler radar to measure ground speed and this radar needs few minutes to warm-up before it can operate. The authover is dependent on this radar and cannot operate without the doppler radar.

That's one sweet model :) I'll almost feel sorry for ruining it with a GAU-8 strafe..oh well.

The principle of track system is very simple. It's just a record of timestamped in-game events and keystrokes packed with the mission itself. So the track playback is done by executing the same in-game events and doing the same key-strokes at the recorded time while running the mission. That's why you cannot rewind a track or quick-jump to a specific time. The track are played at fps the computer can handle, becuase for the simulation engine it's just a normal mission. Having a split functionality would require some kind of save mechanism, that would be able to save a complete mission state, including triggers, AI routines states and every other underlying simulation variables, and that's not a simple task.

You mean that game with free core, with few free planes, and payware addon planes. Sounds familiar to the model ED is adopting with DCS world. also note, that WWI planes are much much more simplier then modern planes, so they are able to produce several new-ones a year, with a lower pricetag. The added complexity in modern planes prevents this.

Ofc, you're right. I was trying to explain why pressing TMS Up long on HUD doesn't reset the SPI back to SP.

TMS up Long while hud is SOI makes SPI at the TDC (the rectangle thingie you slew around the hud) location. To reset SPI back to steerpoint you need to press TMS AFT Long

AFAIK it's the maximal INS deviation, in NM, for one hour of flight.